Hybridizing Old and New World camelids: Camelus dromedarius x Lama guanicoe

Abstract

Thirty female dromedary camels were inseminated on a total of 50 occasions with 2-4 ml of fresh guanaco semen diluted with an equal volume of commercially available camel semen extender. Similarly, nine female guanacos were inseminated on 34 occasions with 4-6 ml of fresh, diluted camel semen. Only two of the dromedary females conceived; one aborted a female foetus on day 260 of gestation and the other gave birth to a stillborn female calf on day 365. Six conceptions occurred in the female guanacos. Two of these conceptuses, diagnosed by ultrasound, were resorbed between days 25 and 40 of gestation, one female foetus was aborted on day 291, another female foetus was aborted on day 302, and one female calf was stillborn on day 365 of gestation. The sixth foetus, a male, was born prematurely but alive after a 328-day gestation. It had a phenotypic appearance intermediate between that of a camel and a guanaco and its hybrid parentage was confirmed by the DNA fingerprinting of eight llama microsatellites. To our knowledge, this is the first viable hybrid ever to be produced between Old World and New World camelids, which have been reproductively isolated from one another for at least 11 million years. The preponderance of female hybrids is in accordance with Haldane's law. Histological examination of their ovaries revealed a failure of meiosis, with only an occasional abnormal oocyte surrounded by follicle cells. Although the diploid chromosone number of camels and guanacos is the same (2n = 74), sufficient genetic change has taken place to make the pairing of homologous chromosomes no longer possible.

Full text

Hybridizing Old and New World camelids:
Camelus dromedarius  Lama guanicoe
J. A. Skidmore1, M. Billah1, M. Binns2, R. V. Short3and W. R. Allen4*
1The Camel Reproduction Centre, PO Box 11808, Dubai, United Arab Emirates
2Animal HealthTrust, Lanwades Park, Newmarket, Su¡olk CB8 7UU, UK
3The Royal Women's Hospital, Carlton, Victoria 3053, Australia
4Department of ClinicalVeterinary Medicine, Equine Fertility Unit, University of Cambridge, Mertoun Paddocks, Woodditton Road,
Newmarket, Su¡olk CB8 9BH, UK
Thirty female dromedary camels were inseminated on a total of 50 occasions with 2^4 ml of fresh
guanaco semen diluted with an equal volume of commercially available camel semen extender. Similarly,
nine female guanacos were inseminated on 34 occasions with 4^6 ml of fresh, diluted camel semen. Only
two of the dromedary females conceived; one aborted a female foetus on day 260 of gestation and the
other gave birth to a stillborn female calf on day 365. Six conceptions occurred in the female guanacos.
Two of these conceptuses, diagnosed by ultrasound, were resorbed between days 25 and 40 of gestation,
one female foetus was aborted on day 291, another female foetus was aborted on day 302, and one female
calf was stillborn on day 365 of gestation. The sixth foetus, a male, was born prematurely but alive after
a 328-day gestation. It had a phenotypic appearance intermediate between that of a camel and a
guanaco and its hybrid parentage was con¢rmed by the DNA ¢ngerprinting of eight llama microsatel-
lites. To our knowledge, this is the ¢rst viable hybrid ever to be produced between Old World and New
World camelids, which have been reproductively isolated from one another for at least 11 million years.
The preponderance of female hybrids is in accordance with Haldane's law. Histological examination of
their ovaries revealed a failure of meiosis, with only an occasional abnormal oocyte surrounded by follicle
cells. Although the diploid chromosone number of camels and guanacos is the same (2n74), su¤cient
genetic change has taken place to make the pairing of homologous chromosomes no longer possible.
Keywords: camel; guanaco; hybrid
1. INTRODUCTION
The family Camelidae is of great antiquity. Palaeontolo-
gical evidence suggests they split o¡ from the other
cloven-hoofed mammals in the Eocene ca. 40^45 million
years (Myr) ago (Romer 1966; Simpson 1980), and split
again into the genera Camelus and Lama in North
America relatively soon after, ca. 30 Myr ago (Webb 1974;
Harrison 1979). However, more recent molecular studies
of mitochondrial DNA (mtDNA) mutation rates suggest
that the split may have occurred ca. 11Myr ago (Stanley
et al. 1994). Today, there remains two species of large, Old
World camels indigenous to Asia and Africa, namely, the
two-humped Bactrian (Camelus bactrianus) and the one-
humped dromedary (Camelus dromedarius), and four species
of smaller New World camels in South America, the
domesticated llama (Lama glama), its probable wild ante-
cedent the guanaco (Lama guanicoe), the domesticated
alpaca (Lama pacos) and its possible wild ancestor the
vicu·a (Lama vicugna). During one of the early ice ages,
the cameloids crossed the land bridge created between
North America and Asia in the region of the Bering
Straits and then radiated through Asia into Eastern
Europe and the Middle East. They were unknown to the
ancient Egyptians (Zeuner 1963), and may have been
introduced into North Africa by man. The lamoids
migrated southwards and crossed the land bridge into
South America. All Camelidae had become extinct in
North America before historic times.
Today, the New and Old World camelids show some
remarkable anatomical and physiological similarities, and
some equally striking contrasts as a result of their adapta-
tion to di¡erent environmental extremes. For example,
they all share the same diploid chromosome number
(2n74; Hsu & Benirschke 1969), they are all induced
ovulators (San-Martin et al. 1968; Musa & Abusineima
1978; El Wishy 1987) and they all have a bicornuate
uterus in which the left horn is considerably larger than
the right and always acts as the site of implantationö
although ovulation occurs equally frequently from the left
and right ovaries (Arthur et al. 1985; El Wishy 1988). They
also share the same di¡use, non-invasive epitheleochorial
placenta (Van Lennep 1963; Steven et al. 1980; Skidmore
et al. 1996a). They di¡er in that the New World camelids
are small, cloven-hoofed, and have a dense, ¢ne-wool
coat that enables them to survive in the extremely low
temperatures of the snowy deserts of the high Andes,
whereas the Old World camelids are much larger, have a
single broad footpad and a less dense hairy coat, and are
adapted to the extreme diurnal temperature variations
Proc. R. Soc. Lond. B (1999) 26 6, 649^656 649 &1999 The Royal Society
Received 2 December 1998 Accepted 21 December 1998
*Author for correspondence.

and scarce food supplies in the mountainous deserts of
Mongolia and the low-lying deserts of Arabia. Both the
Old World camelids will hybridize with one another to
produce fertile o¡spring, and all the New World species
will similarly hybridize with each other and produce
fertile young (Gray 1972). However, to our knowledge
there are no published accounts of induced hybridization
between Old and New World camelids, which could not
occur naturally due to geographical separation and great
disparity in body size. The object of the present study was
to attempt to hybridize dromedaries with guanacos, using
arti¢cial insemination.
2. MATERIALS AND METHODS
Thirty adult female dromedary camels aged 6^14 years and
estimated to weigh between 380 kg and 450 kg, and two adult
male dromedaries aged ¢ve and eight years and estimated to
weigh 550^6 00 kg that had been trained to ejaculate into a
modi¢ed bull arti¢cial vagina (AV), were maintained as part of
the experimental herd at the Camel Reproduction Centre in
Dubai, UAE. Nine females (estimated age, 3^7 years; estimated
weights, 75^80 kg) and one young male (estimated age, 3^4
years; estimated weight, 85kg) guanacos were captured from a
nearby wildlife park, translocated to pens at the Camel Repro-
duction Centre and partly tamed.
During the camel breeding season in the Arabian Gulf region
(November ^April) of 1995^1996 and 1996^1997, the ovarian
follicular wave patterns of the dromedaries were monitored by
serial trans-rectal ultrasound examinations, as described by
Skidmore et al. (1995). When a dominant follicle reached a
diameter of 1.3^1.6 cm the camel was given an intravenous injec-
tion of 20 mg of the GnRH analogue, buserelin (Receptal;
Hoechst Animal Health, Bedfordshire, UK). The camel was
then inseminated with the whole ejaculate from a male guanaco
(2^4 ml of semen; 15^4 0107spermatozoa; 50^70% motility)
collected with the AV and diluted with an equal volume of a
commercial semen extender designed for camel semen (Green
Bu¡er; IMV Ltd, L 'Aigle, France) containing 10% v/v egg
yolk. This was deposited in the body of the uterus using a plastic
bovine insemination catheter guided manually through the
cervix by the operator's sterile gloved hand in the vagina. These
inseminations were given either once, 24 h after treatment with
GnRH (n45), or twice, at the time of GnRH therapy and
again 24 h later (n5).
The female guanacos were also monitored by trans-rectal
ultrasonography and when the dominant follicle attained a
diameter of 0.8^0.9 cm, each guanaco was given an intra-
muscular ( IM) injection of 10 mg buserelin and inseminated
either once 24 h later (n22) or t wice 24 h apart (n11) with a
4^6 ml aliquot of camel semen diluted 1:1 with Green Bu¡er
containing 10% v/v egg yolk. The inseminate was prepared
from a camel ejaculate (3^8 ml; 50^150107spermatozoa; 60^
80% motility) and it was also deposited in the uterine body by
means of a manually guided insemination catheter passed
through the cervix.
Ovulation in the camels and guanacos was diagnosed by
ultrasound examination of the ovaries 48 h after the GnRH
injection (Skidmore et al. 1995) and con¢rmed subsequently by
measuring a rise in progesterone concentrations in peripheral
serum samples recovered daily from each animal from the time
of the GnRH treatment. A chemiluminescent progesterone
assay developed for human serum (Amerlite; Kodak Diagnostics
Ltd, Buckinghamshire, UK) and validated for camel serum by
Skidmore et al. (1996b) was used. Pregnancy was suspected
when serum progesterone concentrations remained elevated
beyond 12 d after insemination, and was subsequently con¢rmed
by ultrasound examination of the uterus. A discrete accumula-
tion of conceptus £uid in the lumen of the left uterine horn was
¢rst observed between days 18 and 20, an echogenic embryo
suspended within the £uid could be distinguished between days
20 and 23, and a foetal heartbeat was identi¢able between days
25 and 27 (Skidmore et al. 1992).
In two of the inseminated camels it was suspected that luteal
function was impaired, as judged, using ultrasonography, by the
small size of the corpus luteum and low serum progesterone
concentrations. These animals were, therefore, given daily IM
injections of 6 ml of a 25 mg ml71suspension of progesterone in
peanut oil (Intervet Laboratories, Cambridge, UK) from the
sixth day after insemination until one of them was con¢rmed as
non-pregnant by ultrasound examination on days 22 and 25 and
the other aborted a dead and partly autolysed foetus on day 260
of gestation.
3. RESULTS
Thirty dromedary females were inseminated on 50
occasions with diluted guanaco semen 24 h after an injec-
tion of GnRH analogue was given to induce ovulation,
but only two conceived (table 1). One camel aborted a
dead and partly autolysed female foetus (GC2) on day
260 of gestation while still receiving progesterone therapy.
The other calved unattended and spontaneously on day
365 of gestation but the female calf was stillborn (GC1).
Post-mortem examination showed that the calf had devel-
oped normally but its lungs had not been aerated.
Six conceptions occurred in the nine female guanacos
that were inseminated with diluted camel semen after a
GnRH injection to induce ovulation on 34 occasions
(table 1). One conceptus failed to develop an embryo and
the vesicle shrank and disappeared ultrasonographically
between 22 and 30 days after insemination (CG3). In a
second guanaco, the foetal heartbeat ceased between the
two serial ultrasound examinations performed 30 and 39
days after insemination, and the remaining conceptus
£uid and membranes shrank and disappeared altogether
over the following 20 days (CG4). One female foetus was
aborted on day 291 (CG1), another female foetus was
aborted on day 302 (CG6) and one female foetus was
stillborn on day 365 of gestation (CG2). The sixth
guanaco calved spontaneously and unaided on day 328 of
gestation and produced a live male calf, Rama (CG5),
that was somewhat premature and weighed only 5.5 kg.
This was less than the weight of a newborn guanaco at
term (8^10 kg) and appreciably less than the weight of a
newborn camel calf ( 30 kg).
(a) Rearing the calf
The hybrid calf showed no tooth eruption at birth, an
indication of its prematurity, and since its mother had no
mammary development and showed no maternal beha-
viour, it was hand-reared on fresh camel milk obtained
daily from a dromedary that had calved 24 h earlier.
Initially, each feed of 30^50 ml of camel milk was o¡ered
at regular intervals, every 1.5 h throughout the day and
night. After 48 h, the feeding interval was extended to 2 h
650 J. A. Skidmore and others Hybrids between camels and guanacos
Proc. R. Soc. Lond. B (1999)

and the daily intake increased to 120 0 ml. This pattern
was continued for the ¢rst month but, during the second
and third months, the feeding interval was increased
gradually to 4 h. The volume of milk consumed per feed
increased from 300 ml at the end of the ¢rst month to
700 ml by the end of the fourth month, by which time the
feeding interval had extended to 8h. A compounded `calf
starter' feed and hay were o¡ered to the calf from
30 days of age and its daily intake of these solids had
increased to around 500 g by the fourth month. Initially,
the calf gained weight at a rate of 0.2 kg per day over the
¢rst seven days but thereafter the gain increased to
approximately 2.5 kg per week (¢gure 1).
At nine months of age the hybrid calf was in good
health. He exhibited the woolly ¢bre coat and the nose
and nostrils of the New World camelids, but his ears and
tail were midway in length between those of camels and
guanacos (¢gure 2c). Similarly, his feet were an inter-
mediate of the single two-toed conjoined footpad of
camels and the cloven hooves of guanacos (¢gures 2d^f).
However, unlike guanacos, he showed no skin glands on
the lateral or medial aspects of the tarsus and there was
no sign of the hump that would be present on a camel calf
of the same age. He had two small testicles, measuring
4 cm 2cm and palpable in the scrotum ca. 4 cm below
the anus.
Hybrids between camels and guanacos J. A. Skidmore and others 651
Proc. R. Soc. Lond. B (1999)
Table 1. Hybrid pregnancies
(C represents camel; G represents guanaco.)
hybrid
number sire dam
date of
delivery
foetal weight
(kg) sex
duration of
gestation (days)
camel guanaco
CG1 Musehan 3 3/11/96 2.5 female 291 (aborted)
CG2 Musehan 3 4/1/98 9.0 female 365 (stillborn)
CG3 Young One or Musehan 1 ööö30 (resorbed)
CG4 Young One or Musehan 4 ööö40 (resorbed)
CG5 Musehan or Young One 1 14/1/98 5.5 male 328 (born live)
CG6 Musehan 6 3/11/98 3.7 female 302 (aborted)
guanaco camel
GC1 Whalid 660 18/2/97 30.0 female 365 (stillborn)
GC2 Whalid 1610 3/2/98 1.0 female 260 (aborted)
Figure 1. Growth curve for the male dromedary guanaco calf, Rama, during the ¢rst nine months. Open circles, height in
centimetres; closed circles, weight in kilograms.

652 J. A. Skidmore and others Hybrids between camels and guanacos
Proc. R. Soc. Lond. B (1999)
Figure 2. (a) The dromedary sire, Musehan, and the guanaco dam (no. 1) with their two-day old hybrid calf. (b,c) The hybrid
calf, Rama, at two days and two months of age, respectively. (d^f) Comparison of the front footpads of (d) an adult female
guanaco, (e) the dromedaryguanaco hybri d, and ( f) a newborn dromedary camel calf.

Hybrids between camels and guanacos J. A. Skidmore and others 653
Proc. R. Soc. Lond. B (1999)
Figure 3. Histological sections of the ovaries recovered from (a) a full-term guanaco calf and (b^d) three guanacodromedary
hybrid foetuses (see table 1). (a) The guanaco guanaco calf stillborn at full term. The cortex is packed with `normal'-looking
oocytes contained within primary, secondary or tertiary follicles. (b) Hybrid calf CG2, showing a compact, ¢brous cortex
containing many eosinophylic cords which may represent dead or dying oogonia in P£u
«ger's cords, unable to enter meiosis. No
oocytes or follicles are visible. (c) A high-power section of the cortical region of the ovary from hybrid GC1 showing heavily
eosinophylic oogonia arranged in clumps or cords, but no oocytes or follicle cells. (d) A high-power section of the ovary from
hybrid CG1 showing one unhealthy looking oocyte surrounded by degenerating follicle cells and an adjacent, even more
degenerative, oocyte with a pycnotic nucleus and barely recognizable follicle cells. Scale bars, 100 mm.

Behaviourally, Rama clearly showed features of his
hybrid genotype. His vocalization was an unusual highly
pitched `croak' uttered on expiration, which has discern-
ible elements of both parental species. He urinated
backwards in a series of spurts like both guanacos and
camels, but defecated while moving which is similar to
camels but not guanacos, which tend to defecate in one
place. He chewed cud by alternating from one side to the
other in a manner similar to guanacos. He developed
aggressive behaviour towards adult female guanacos,
which included laying his ears back, rearing onto his hind
legs, striking out with his front feet and attempting to
spit; none of these movements are exhibited by young
camels.
(b) Post-mortem ¢ndings
The ovaries of hybrid foetus GC2 were too decom-
posed to permit histological examination. Ovaries were
recovered from the hybrid foetus, CG1, aborted on day
291 of gestation; the two hybrid calves, CG2 and GC1,
stillborn on day 365; from a newborn full-term guanaco
calf; and from a dromedary foetus aborted at 12 months
of gestation. Pieces of ovarian tissue were ¢xed in 10%
phosphate-bu¡ered formaldehyde solution, embedded in
para¤n wax, sectioned at 5 mm thickness and stained
with haematoxylin and eosin (H & E).
Histologically, ovaries from the full-term guanaco
revealed a number of tertiary Graa¢an follicles with a
£uid-¢lled antrum, and many primordial and secondary
follicles throughout the outer cortex, with oocytes
surrounded by single or multiple layers of follicle cells
(¢gure 3a). The ovaries of the camel foetus contained
numerous large tertiary follicles, of up to 0.5 cm in
diameter, and many primordial and secondary follicles
throughout the outer ovarian cortex.
The ovaries of the three hybrids presented a very
di¡erent appearance. GC1, stillborn at 365 days, showed
heavily eosinophilic oogonia arranged in clumps or cords,
but there was no evidence that any of the oogonia had
entered meiosis as there were no signs of any oocytes with
their characteristic layer of surrounding follicle cells
(¢gure 3c). CG2, also stillborn at 365 days, showed a
compact, ¢brous ovarian cortex with many eosinophilic
cords which may represent dead or dying oogonia that
were unable to enter meiosis (¢gure 3b). There were
occasional degenerating oocytes with surrounding follicle
cells. The ovaries of CG1, aborted at 291 days, showed a
similar picture, with an occasional degenerate-looking
oocyte surrounded by degenerating follicle cells,
suggesting that a few oogonia had been able to initiate
meiosis. However, chromosomal pairing was probably
incomplete and resulted in the death of the oocyte and
follicle cells that it had induced to form around itself
(¢gure 3d).
(c) Parentage analysis
Blood samples were collected into preservative-free
sodium heparin tubes from the two male camels,
Musehan and Young One, that had supplied the semen
used for arti¢cial insemination; from guanaco no. 1, the
mother of the calf; and from Rama the calf (CG5, see
table 1). DNA was extracted from a 5 ml aliquot of whole
blood using a commercial kit (Nucleon; Scotlab Ltd,
Coatbridge, Scotland). Subsequently, genotypes for each
animal were obtained by performing PCR on the
extracted DNA, using eight llama microsatellites under
the PCR conditions described by Lang et al. (1996).
Fluorescent dUTP was incorporated into the PCR
products which were then electrophoresed on an ABI377
automated sequencer and the data analysed using
Genescan 2.1.
The results of the genotyping analysis using the llama
microsatellites are presented in table 2. They are fully
consistent with the hybrid being the o¡spring of the
female guanaco and the male camel, Musehan, with
exclusion of the second male camel, Young One, on the
basis of two of the markers,YWLL08 and YWLL44. The
size ranges for the alleles ampli¢ed from the guanaco
sample were all within the ranges published by Lang et al.
(1996), except those of markers YWLL19 and YWLL44.
The variation observed with YWLL44 was relatively
small, the observed alleles being 83/112, compared with
the published range of 86^120. However, for YWLL19,
the size of the observed allele (243) di¡ered signi¢cantly
from the published size range of 137^161.
4. DISCUSSION
To our knowledge, this is the ¢rst report of a viable
hybrid between Old and New World camelids. It was
achieved by using arti¢cial insemination and hormone
therapy to overcome the marked di¡erences between the
two parental species in terms of their body size and their
oestrous behaviour (England et al. 1971; Skidmore et al.
1996b). Considering the other marked anatomical,
physiological and behavioural di¡erences that have
evolved between the two separated groups to enable them
to survive in very harsh and diametrically opposed
enviroments, it is quite remarkable that their basic repro-
ductive mechanisms have been su¤ciently conserved to
permit hybridization. Although the diploid chromosome
number has remained unchanged at 2n74 for all
camelids, the failure of meiosis in the ovaries of all the
hybrids suggests that su¤cient genetic change must have
654 J. A. Skidmore and others Hybrids between camels and guanacos
Proc. R. Soc. Lond. B (1999)
Table 2. Genotyping results for the eight llama microsatellites tested on the camel guanaco hybrid and its parents
YWLL08 YWLL19 YWLL29 YWLL36 YWLL40 YWLL43 YWLL44 YWLL59
guanaco (no. 1) 135/143 243/243 215/217 153/155 186/186 133/133 83/112 107/111
hybrid (CG5) 135/158 243/243 205/215 133/155 172/186 133/133 83/103 107/107
Musehan 158/158 243/243 205/205 131/133 172/172 133/133 103/107 107/109
Young One 128/130 243/243 205/205 131/133 172/172 133/133 107/107 107/109

occurred during the millions of years of reproductive
isolation to disrupt the pairing of homologous chromo-
somes as they enter meiosis. It remains to be seen whether
the surviving male hybrid will be capable of producing
any sperm when he reaches puberty, although this seems
most unlikely.
The conception rates achieved when inseminating
dromedaries with semen from a single male guanaco (two
conceptions from the insemination of 30 fertile females on
50 separate occasions) and guanacos with semen from
two dromedary males of known fertility (six conceptions
from the insemination of nine animals on 34 occasions)
were much lower than the 50^55% conception rate
achieved routinely in the same laboratory when insemi-
nating dromedaries with dromedary semen. This might
indicate impaired fertilization and/or early embryonic
development. The conception rate appeared to be higher
in guanacos inseminated with dromedary semen than in
the dromedaries inseminated with guanaco semen, which
is reminiscent of the di¡erence in fertility between other
reciprocal interspeci¢c mammalian matings, such as
between rabbit and hare (Chang et al. 1964), sheep and
goat (Hancock et al. 1968), and horse and donkey (Allen
& Short 1997). It could be explained by di¡erences in
sperm^egg binding; camel spermatozoa might bind more
readily to guanaco oocytes than vice versa. The causes of
the late foetal deaths and stillbirths in the camelid
hybrids remain unexplained.
DNA genotyping was used to con¢rm that the hybrid
was the product of the mating between the female
guanaco and a male dromedary camel and it also deter-
mined the actual sire between the two possible conten-
ders. The combined exclusion probabilities of the eight
microsatellites used, based on their individual exclusion
probabilities in llamas and alpacas, is 0.9997. While the
corresponding exclusion probability in Old World came-
lids has not been determined, it would be surprising if
this panel of microsatellites did not display useful poly-
morphism in both the Bactrian and dromedary camels.
The allele size observed with the marker YWLL19 in all
four animals typed here (243 bp) di¡ers signi¢cantly
from the published allele size range in llamas and alpacas
(137^161bp). The primer sequences for YWLL19 were
veri¢ed as correct and no obvious explanation exists for
this discrepancy. It was of interest to ¢nd that the allele
sizes observed with the camel samples for ¢ve of the
markers (YWLL08, YWLL19, YWLL29, YWLL36 and
YWLL40) fell outside the ranges for these markers
observed in llamas and alpacas, thereby suggesting that
these microsatellites have diverged considerably in Old
and New World camelids since their split 11^30 Myr ago.
Perhaps one of the most interesting aspects of this study
relates to the gestation length and size at birth of the
surviving hybrid. The duration of gestation (328 days)
and birth weight (5.5 kg) are close to the normal range
for guanaco pregnancy, but are signi¢cantly less than the
395 days 30 kg of a normal dromedary pregnancy.
Thus, there appears to have been a complete maternal
override of the paternal genotype during pregnancy,
which is perhaps a result of the size constraint imposed
by a restricted area of the endometrium upon the di¡use,
non-invasive epitheliochorial placenta. Once born,
however, the hybrid calf had shown signi¢cant catch-up
growth and at nine months of age was 2.5 cm taller than
his mother. It will be interesting to see whether this
pronounced intra-uterine growth retardation has any
adverse e¡ects on his subsequent health and well-being,
as the Barker hypothesis (Barker 1995) would predict.
The apparent female skewing of the sex ratio in the
foetuses and neonate (one male:¢ve females) is in accor-
dance with Haldane's law, which states that `when in the
F1o¡spring of two di¡erent animal species one sex is
absent, rare or sterile, that sex is the heterozygous sex'
(Haldane 1922). The reason for this skewing may be that
the mutation rate of genes of the unpaired segment of the
Y chromosome is apparently much higher than that of
genes on any other chromosome (Short 1997a). This is
because any genetic defects cannot be `repaired' by
meiotic crossing over with a homologous chromosome.
Furthermore, the Ychromosome never enters female germ
cells, and germ-line mutations are known to be far more
common in the testis than the ovary (Short 1997a,b).
Considering the many millions of years of reproductive
isolation of the dromedary and the guanaco, it is perhaps
surprising that any male hybrids were produced at all.
Rama is living proof that the sex-determining genes on
the camel Y chromosome are still capable of inducing
testicular development in the hybrid. However, it seems
likely that Rama will prove to be sterile, both because
of the generalized meiotic failure observed in the
ovaries of the hybrids, and the expected di¡erences
between Y-linked spermatogenesis-determining genes in
the dromedary and the guanaco.
This study was kindly sponsored by H. H. Sheikh Mohammed
bin Rashid al Maktoum, Crown Prince of Dubai. We wish to
thank Professor Ulrich Wernery for his help with post-mortem
examinations, Bruce Abaloz for assistance with histology and
David Paul for the photomicrography. Dr K. Benirschke kindly
provided sections of the newborn guanaco ovary.
REFERENCES
Allen, W. R. & Short, R. V. 1997 Interspeci¢c and extra-
speci¢c pregnancies in equids: anything goes. J. Hered. 88,
384^392.
Arthur, G. H., Rahim, A. T. A. & Hindi, A. S. A. 1985
Reproduction and genital diseases of the camel. Br.Vet. J. 141,
650^659.
Barker, D. J. P. 1995 The foetal origins of adult disease. Proc. R.
Soc. Lond. B262, 37^43.
Chang, M. C., Marston, J. H. & Hunt, D. M. 1964 Reciprocal
fertilisation between the domesticated rabbit and the snow-
shoe hare with special reference to insemination of rabbits
with an equal number of hare and rabbit spermatozoa. J. Exp.
Zool. 155, 437^445.
El Wishy, A. B. 1987 Reproduction in the female dromedary
(Camelus dromedarius): a review. Anim. Reprod. Sci.82, 587^593.
El Wishy, A. B. 1988 A study of the genital organs of the
female dromedary (Camelus dromedarius). J. Reprod. Fert. 82,
587^593.
England, B. G., Foote, W. C., Cardozo, A. G., Matthews, D. H.
& Riera, S. 1971 Oestrous and mating behaviour in the llama
(Llama glama). Anim. Behav. 19, 722^726.
Gray, A. P. 1972 Mammalian hybrids. Farnham, Buckinghamshire:
Commonwealth Agricultural Bureaux.
Haldane, J. B. S. 1922 Sex ratio and unisexual sterility in hybrid
animals. J. Genet. 12, 101^109.
Hybrids between camels and guanacos J. A. Skidmore and others 655
Proc. R. Soc. Lond. B (1999)

Hancock, J. L., McGovern, P. T. & Stamp, J. T. 1968 Failure of
gestation of goat sheep hybrids in goats and sheep. J. Reprod.
Fert. 3(Suppl.), 29^36.
Harrison, J. A. 1979 Revision of the Camelidae (Artiodactyla,
Tylopoda) and description of the new genus Alforjas. Paleontol.
Contrib. (University of Kansas) 95, 1^27.
Hsu, T. C. & Benirschke, K. 1969 Atlas of mammalian cytogenetics.
NewYork: Springer.
Lang, K. D. M., Wang, Y. & Plante, Y. 1996 Fifteen polymorphic
dinucleotide microsatellites in llamas and alpacas. Anim.
Genet. 27, 293^297.
Musa, B. E. & Abusineima, M. E. 1978 The oestrous cycle of
the camel (Camelus dromedarius). Vet. Rec. 102, 556^557.
Romer, A. S. 1966 Vertebrate paleontology, 3rd edn. University of
Chicago Press.
San-Martin, M., Copairo, M., Zuniga, J., Rodreguez, R.,
Bustinza, G. & Acosta, L.1968 Aspects of reproduction in the
alpaca. J. Reprod. Fert. 16, 395^399.
Short, R. V. 1997aThe testis: the witness of the mating system,
the site of mutation and the engine of desire. Acta Paediatr.
422 (Suppl.), 3^7.
Short, R. V. 1997bAn introduction to mammalian interspeci¢c
hybrids. J. Hered. 88, 355^357.
Simpson, G. G. 19 80 Splendid isolation. Yale University Press.
Skidmore, J. A., Billah, M. & Allen, W. R. 1992 Ultra-
sonographic and videoendoscopic monitoring of early fetal
development in the dromedary camel. In Proceedings of the First
International Camel Conference (ed. W. R. Allen, A. J. Higgins,
I. G. Mayhew, D. H. Snow & J. F. Wade), pp. 193^201.
Newmarket, UK: R. & W. Publications.
Skidmore, J. A., Billah, M. & Allen, W. R. 1995 The ovarian
follicular wave pattern in the mated and non-mated
dromedary camel (Camelus dromedarius). J. Reprod. Fert. 49
(Suppl.), 545^548.
Skidmore, J. A., Wooding, F. B. P. & Allen, W. R. 1996a
Implantation and early placentation in the one-humped
camel (Camelus dromedarius). Placenta 17, 253^262.
Skidmore, J. A., Billah, M. & Allen, W. R. 1996bThe ovarian
follicular wave pattern and induction of ovulation in the
mated and non-mated one-humped camel (Camelus drome-
darius). J. Reprod. Fert.10 6, 185^192.
Stanley, H. F., Kadwell, M. & Wheeler, J. C. 1994 Molecular
evolution of the family Camelidae: a mitochondrial DNA
study. Proc. R. Soc. Lond. B25 6, 1^6.
Steven, D. H., Burton, G. J., Sumah, J. & Nathanielsz, P. W.
1980 Ultrastructural observations on the placenta of the
alpaca, Lama pacos.Placenta 1, 21^32.
Van Lennep, E. W. 1963 The histology of the placenta of the
one-humped camel (Camelus dromedarius) during the second
half of pregnancy. Acta Morphologia Neerlandica Scandinavia 6,
373^379.
Webb, S. D. 1974 Pleistocene mammals of Florida. University of
Florida Press.
Zeuner, F. E. 1963 A history of domesticated animals. London:
Hutchinson.
As this paper exceeds the maximum length normally permitted,
the authors have agreed to contribute to production costs.
656 J. A. Skidmore and others Hybrids between camels and guanacos
Proc. R. Soc. Lond. B (1999)